1. Field of the Invention
The present invention relates to a method for resolution conversion and, more particularly, to a method of conversion of a facsimile-based resolution to a printer-based resolution in the vertical direction.
2. Description of the Prior Art
Digital image processing technology has been being pushed by the incessant demands from the highly developed electronic devices, such as scanners, printers, facsimiles, digital cameras, and the like. The development of digital image process makes pictorial webs possible and further facilitates people in managing image files or documents both through the Internet and on office work. Printers are well-known devices for producing copies from PCs, facsimile or fax machines, and typically, image signals used for facsimiles and printers are respectively formatted on a pixel/mm and dot-per-inch (dpi) basis. In making a copy based on image signals of a facsimile via a printer engine, it is inevitable to sacrifice some image signals or to duplicate some image signals due to the different formats employed.
For an 8xc3x973.85 pixels/mm, which is equivalent of 203.2xc3x9797.8 dpi, of a formatted facsimile signal, both the vertical and horizontal resolutions need to be converted to, for example, 300 dpi. For this, duplication must be adopted to achieve the resolution conversion. In a prior art, 203.2xc3x9797.8 dpi is converted to 300xc3x97300 dpi by two stages. In the first stage, 203.2xc3x9797.8 is converted to 200xc3x97100 dpi, wherein duplication is used; and then converted to 300xc3x97300 dpi referred to as the second stage. In the first stage, two lines, for example, the 49th and 98th lines are added for every 98 lines in the vertical direction by means of software conventionally, and 3 dots are deleted in the horizontal direction from every 203 dots so that a resolution of 200xc3x97100 dpi is achieved. When 100 dpi is converted to 300 dpi in the second stage via hardware implementation such as an ASIC, there need some duplications in the vertical direction. To map the 100 lines in an inch range into 300 lines per inch, each of the 100 lines needs to be duplicated for two times. Consequently, the 49th line and the 98th line of the 98 lines in an inch are respectively drawn for six times, but other lines are just drawn for 3 times. FIG. 1 shows the aforementioned two stage conversions, wherein the original image 102 is generated from a facsimile machine, and the images 104 and 106 are respectively derived from the first and second stages. Therefore, in this scheme for conversion of vertical resolution from a facsimile-based signal format to a printer-based signal format, the errors are significantly amplified and hence a non-realistic image is provided. Worse yet, the errors are oddly distributed throughout a piece of paper.
The principal object of the present invention is to disclose a method of resolution conversion for more evenly distributed errors in the vertical direction.
According to the above object, the first embodiment of the disclosed method divides the printer-based vertical resolution by the facsimile-based vertical resolution to obtain a quotient and a remainder, both being integers. Some data lines of the printing job formatted on the facsimile-based resolution, in which their total number is equal to the remainder, are selected for the purpose of being generated one more time than the others. All the data lines are then repeatedly generated for xe2x80x9cquotientxe2x80x9d times except those selected ones, which are generated for xe2x80x9cquotient+1xe2x80x9d times. The selected data lines can be randomly or fixedly selected.
According to the second embodiment, the disclosed method also divides the printer-based vertical resolution by the facsimile-based vertical resolution to obtain a quotient and a remainder both integers, too. All the data lines are repeatedly generated for xe2x80x9cquotientxe2x80x9d times. Next, some data lines that their total number equals to the remainder are selected from the repeated data lines in order to be repeated once more than the other ones. As same as the first embodiment, the selected data lines can be randomly or fixedly selected from the repeated data lines.